01
Transaction monitoring & AML operations
Rule design, threshold tuning, typology coverage, alert quality and analyst QA. The unglamorous work that determines whether a programme detects anything at all — and whether it can prove it.
Sulaiman Tidjani — Financial Crime Compliance & Applied AI
Compliance that keeps pace with financial crime.
Where anti-financial-crime meets applied artificial intelligence — monitoring, screening and fraud controls designed to detect more of what matters, with fewer false positives and a record that stands up to a regulator.
Risk aware. Compliance assured. Value delivered.
AML
Financial Compliance
AI Systems
Sanctions Screening
Model Governance
01 — Profile
A practitioner's view, not a vendor's.
Most of what is written about AI in compliance is written by people selling AI to compliance teams. I write from the other side of the desk — from inside the alert queue, the QA sample, the tuning debate, and the conversation where someone has to explain to a regulator why the model did what it did.
My work sits across transaction monitoring, sanctions and fraud risk — threshold and typology design, quality-assurance frameworks for investigation decisions, and assessing where machine learning genuinely outperforms a well-tuned rule, and where it quietly does not.
This site publishes general commentary on AI in fraud prevention and financial crime detection — drawn from public regulatory developments, published guidance and industry research. Written in a personal capacity, and containing no confidential or employer-specific information.
02 — Focus
Four areas I work in and write about.
02
AI & machine learning in the control environment
Where models add real lift over rules, and where they add risk. Explainability, model risk management, validation evidence, human-in-the-loop design, and the governance a supervisor will ask for before they ask about performance.
03
Fraud, payments & disputes
Account takeover, authorised push payment scams, mule networks and card-not-present fraud — and the dispute and chargeback controls that keep loss rates and scheme ratios inside a defensible range.
04
Regulatory change, read operationally
FATF guidance, the EU AML package, FinCEN and OFAC developments, and emerging AI regulation — translated out of consultation language and into what has to change in your policies, systems and staffing.
03 — Insights
Selected writing.
Analysis on compliance, financial crime and AI — published regularly.
What AI actually fixes in transaction monitoring — and what it doesn't
Machine learning is very good at ranking alerts and very bad at inventing typologies nobody has described. A practitioner's map of where the lift is real, where it is marketing, and what has to be in place before either.
The EU AML package: what actually lands on operations teams
Beyond the headline of a single rulebook and a new authority — the specific changes to beneficial ownership, cash limits and supervisory expectations that will reshape day-to-day control design.
Measuring an analyst, not their output
Alert throughput is a terrible quality metric. A weighted QA framework for scoring decision quality — narrative, escalation judgement, evidence handling — without turning review into a productivity race.
Keeping dispute rates low: a control view of chargebacks
Chargeback rate is a lagging indicator of decisions made ninety days earlier. Separating true fraud from service failure from cardholder misuse — and what actually moves the number.
A control you cannot explain is not a control. It is an assumption with a dashboard attached.
04 — Advisory
Where I can help directly.
Short, scoped engagements for compliance and risk teams.
Engagement 01
Programme diagnostic
An independent read on how your monitoring and screening estate is actually performing, and the three changes that would move the needle most.
- Coverage and typology gap analysis
- Threshold and tuning review
- Alert-to-outcome funnel assessment
- Prioritised remediation roadmap
Engagement 02
AI readiness & model governance
Before you buy or build: whether your data, controls and documentation can support a model — and the governance you will need to defend it.
- Use-case triage: model vs. rule
- Data and label-quality review
- Explainability and validation evidence
- Vendor due-diligence question set
Engagement 03
Analyst QA & performance frameworks
A weighted scoring model for investigation quality that your team accepts and your second line can rely on.
- Weighted QA category design
- Sampling methodology
- Scorecards and calibration sessions
- Coaching and escalation pathways
05 — Speaking
Talks, panels & internal sessions.
Topics I speak on
- AI in financial crime detection: signal, noise and governance
- Keynote
- Why false positives are a design choice, not a fact of life
- Talk
- Building QA that measures judgement, not throughput
- Workshop
- Regulatory change without programme chaos
- Panel
- Financial crime risk on platforms and marketplaces
- Talk
I speak to compliance functions, RegTech audiences and internal risk teams — conference keynotes and panels, private briefings, and hands-on workshops for monitoring and investigations teams.
Press kit, headshot and full speaker bio available on request.
06 — Contact
Advisory, speaking, or a second opinion.
hello@sulaimantidjani.comAlso on LinkedIn. For media and speaking enquiries, please include date, audience and format.
The Compliance Signal
A short note when something in financial crime compliance or AI actually matters. No roundups, no vendor content.
Occasional. Unsubscribe any time.